Time, Tense and Aspect in Kafka's "The Burrow"

Author(s): J. M. Coetzee
Source: MLN, Vol. 96, No. 3, German Issue (Apr., 1981), pp. 556-579
Published by: The Johns Hopkins University Press
Stable URL: http://www.jstor.org/stable/2905935
Accessed: 20/09/2010 14:49

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=jhup.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

The Johns Hopkins University Press is collaborating with JSTOR to digitize, preserve and extend access to
MLN.

http://www.jstor.org
 Time,Tense and Aspectin Kafka's
"The Burrow"

J. M. Coetzee

Kafka's story"The Burrow" begins: "I have completed the con-

structionof myburrowand itseems to be successful."'The position
in time of the speaker, the creaturewhose lifehas been devoted to
the buildingof thisperfectlysecure hideaway,seems to be clear: he
speaks (or writes)froma momentafterthe completionof the bur-
row but not so long afterit thatfinaljudgment on itssuccess can be
given. Furtherinformationin the nextfewpages help to situatethe
fictionalnowof his utteranceas belongingto "the zenithof mylife"
(p. 325), when he is nevertheless"growingold" (p. 326), "gettingon
in years" (p. 327).
The timeencompassed byhis act of storytelling, beginningat this
is
moment, not, however, simply the time that mightbe taken to
utterthe thirty-five or so pages of the text: although there are no
typographicalbreaks to markbreaks in the timeof narration,there

1The
CompleteStories,ed. Nathan Glatzer (New York: Schocken, 1946), p. 325.
The translationis by Willa and Edwin Muir. Because the Muir translationis the
standard one, I use it throughoutin this essay except at points where the Muirs,
perhaps baffledby Kafka's unusual tensesequences, attemptto smoothout the time
structure by silent emendation. All departures from the Muir translation are
marked by footnotes.The German textused is thatedited byJ. M. S. Pasley in Der
Heizer.In derStrafkolonie.
Der Bau (Cambridge: Cambridge U.P., 1966). Pasley'stext
is based on a freshreading of Kafka's manuscriptand improveson the textgivenby
Max Brod in Franz Kafka,Gesammelte V (New York: Schocken, 1946). For
Schriften,
a cautionaryword about Pasley'stext,however,see Heinrich Henel, "Das Ende von
KafkasDer Bau", Germanisch-Romanische 22 (1972), 22-23.
Monatsschrift,

MLN Vol. 96 Pp. 556-579

0026-7910/81/0963-556 $01.00 ? 1981 by The Johns Hopkins UniversityPress
 M L N 557

is at least one point (p. 343) where the narrationis interruptedby

sleep. As for the time depicted by the narrative,all I shall say as a
firstapproximation is that, aside from passing references to a
farofftime of apprenticeship(e.g., p. 357), it appears to cover life
in the burrow (which it depicts largelyas dominated by habit), to
include and pass beyond the momentat whichthe firstwords of the
textare uttered,and to continue as far as the momentat whichthe
last words are uttered,a moment at which the time of narration
and the time of the narrativeare identical.
But the relationsbetweenthetimeofnarration(the movingnowof
the narrator'sutterance) and the timeof thenarrative(referential
time) turn out to be far more complex and indeed baffling,the
more closelywe read the text.The firstapproximationto a reading
of time-relationsI give above glosses over the problemof fittingthe
patternof habitual life in the burrow into a temporal continuum;
and attemptsto refinethe approximationbring us face to face in
the end withnot onlya narrativestructurebut also a representation
of time which cannot be compressed into a rational model. There
are numerous passages in Kafka's fictionalworks and notebooks
thatreveal a preoccupationwiththe metaphysicsof time.It is above
all in the stories"The Country Doctor" and "The Burrow", how-
ever,thatwe have representations of an idiosyncraticfeel fortime.As
we mightexpect, such storiesnecessarilybring Kafka into conflict
not only withthe time-conventionsof fictionalrealism (which rest
on a Newtonian metaphysics)but also withthe conception embed-
ded in (and, in the Whorfianview,propagated by) the tense-system
of his language.
In thisessay I am concerned to explore the relationsbetweenthe
verb-systemof German (which, in the features I shall be com-
mentingon, is veryclose to the verb-system of English),the narra-
tive (and narratorial)structureof "The Burrow", and the concep-
tionof timewe can postulateKafka held in 1923. In the firstpartof
the essay I attemptlittlemore than to persuade the reader thatthe
taskof layingout the eventsof the narrativein sequential temporal
order is riddled with difficulties.In part II I discuss the work of
two scholarswho have recognized these difficultiesand attempted
to overcome them. In part III I outline a distinctionbetween two
featuresof the verb,tense and aspect,thatare oftenconfused,and
suggesthow upholding the distinctionmay aid us in our reading.
And in part IV I attemptto explain the time-schemethat "The
Burrow" represents,in both senses of this ambiguous phrase.
558 J.M. COETZEE

There is nothing in the firstthree long paragraphs of the text to

conflict with the time and tense conventions of retrospective first-
person narration. But with the fourth paragraph it begins to be-
come more difficult to situate the now of the act of narration in
time. Let us take up this paragraph in some detail.
In the Castle Keep I assemble mystores. . . The place is so spacious that
... I can divide up my stores,walk about among them, play withthem
... That done, I can always... make mycalculationsand huntingplans
for the future,taking into account the season of the year. There are
timeswhen I am so well provided for thatin my indifferenceto food I
never even touch the smaller fry that scuttle about the burrow ...
(p. 328).
The present here is an iterative, habitual present, with a cycle of
seasons and even years.
... It sometimesseems riskyto make the Castle Keep the basis of de-
fense ... Thereupon I mark off every third room ... as a reserve
storeroom ... or I ignore certain passages altogether ... or I choose
quite at random a veryfew rooms . .. Each of these new plans involves
of course heavywork ... True, I can do [it]at myleisure ... But it is not
so pleasant when, as sometimeshappens, you suddenly fancy,starting
up fromyour sleep, thatthe presentdistributionof your storesis com-
pletely and totally wrong . . . and must be set right at once, no matter
how tired or sleepy you may be; then I rush, then I fly,then I have no
time for calculation; and although I was about to execute a perfectly
new, perfectlyexact plan, I now seize whatevermy teeth hit upon and
drag or carry it away, sighing, groaning, stumbling. . . Until little by
littlefullwakefulnesssobers me, and I can ... returnto myrestingplace
... (p. 329).

There is no question that this episode too is iterative, typical, re-

current, and that the now out of which the narrative is uttered is
situated within these recurrences: episodes of panic are part of the
life of the creature, they have occurred in the past, they are ex-
pected to recur.
Then again thereare timeswhen the storingof all myfood in one place
seems the best plan of all ... and so ... I begin once more to haul all my
stores back ... to the Castle Keep. For some time afterwards I find a
certain comfort in having all the passages and rooms free . . . Then I
usually enjoy periods of particulartranquility. . . until at last I can no
longer restrainmyself(bisich es nichtmehrertrage)and one night (eines
 M L N 559

Nachts)rushintotheCastleKeep,mightily upon mystores,

flingmyself
and glutmyself. . . (pp. 329-331).
Here we see thatnarrativewithdifficulty sustainsthe illusionof an
iterativepresent when the actions that recur are impulsive, un-
foreseen,and unforeseeable,when the speaker is at the mercyof
forces he cannot control or predict. Thus the followingsentence
strikesus as bizarre and perhaps ungrammatical:
(1) Every month I impulsivelyrun about the streetsnaked.
Contrastit with:
(2) Every month I run about the streetsnaked.
The only way to domesticate(1) is to read it as a generalization
about behavior over past months,culminatingin the present mo-
ment at whichthe sentence is uttered("Every monthfor the past x
monthsI have impulsivelyrun about the streetsnaked"). It is most
bizarre when it is read as utteredwithinan iterativepresent ("My
habit is impulsivelyto run about the streetsnaked everymonth").
The cause of conflictis of course that for a speaker to take up his
stance withinan iterativepresent means, to the listenerwho, so to
speak, unrolls the cycleof the iterativeon to a past-present-future
continuum,thatthe speaker not onlymakes a generalizationabout
his past behavior but also predictshis futurebehavior; and the act
of predictionconflictswiththe notion of the impulsive.
Kafka does not unequivocally provoke this contradictionin the
passages I have quoted. Nevertheless,both when the burrowing
creature startsout of his sleep and rushes and flies (eile,fliege) to
relocatehis provisions,and when "one night"he rushes (stiirze)into
the Castle Keep to gluthimself,the verbscarryconnotationsof the
impulsive,the uncontrollable,the unpredictable,and thereforesit
uneasily in a narratorialframeworkof iteratedtime.
There are two alternativeways of explaining what is going on
here. The less radical explanation is this: German, like English,
lacks a specificmorphologicalformto signifyiterativeaction. The
non-iterative(punctual) sense of the verb is the semanticallyun-
marked form,in contrastto the marked formof the iterativesense.
(This is perhaps no more than a consequence of the relativeinfre-
quency of the iterativesense.) Therefore unless a sequence of verbs
is systematicallyinterspersed with iterative modifiers (sometimes,
everyday,...) or (in English) is given with an appropriate modal
(will,usedto,. . .), the verbs tend to be read as unmarked, i.e. non-
iterative.In other words, it requires a continual pressure of em-
phasis in the writingto maintainiterativetime.Of course, the more
560 J.M. COETZEE

thisemphasis has to be repeated, the clumsierit sounds. So rather

than maintain the emphasis throughout, Kafka sometimes (for
example, in the last two passages quoted) dramatizes a typicalevent
fromthe iterativecycleand so permitsthe reading to slip back fora
while into the unmarked, non-iterativemode.
This rhetoricalexplanation thus interprets the problematic
verb-sequencesin termsof the pragmaticsof "what works"forthe
reader, as manifestationsof the writer'sartfulness.There is no
doubt that this explanation can be "made to work" for the se-
quences I quote and for others I mentionbelow. My reservations
about explanation along these lines willbecome clearer later in this
essay, when I argue that,ratherthan being an obstacle to under-
standing,the problematicsequences embody a conception of time
thatis centralto Kafka's enterprise.For the momentlet me simply
observe that,"success" in writing,like beauty,being essentiallyun-
demonstrable,it requires some rhetoricalcoaxing and/orintimida-
tion from the critichimselfto establishany argument that a par-
ticular strategyin a text "works", that it is "successful writing",
indeed that it is a "strategyof writing"at all.
The second and more radical explanation is that the time-
conceptionthatreignsin "The Burrow"is trulyaberrant,thatitcan
be domesticated only with a degree of rhetorical violence that
amounts to traduction,and that it is betterunderstood as the re-
flectionof a time-sensewhichdoes not draw a line betweeniterative
and non-iterativesenses of the verb, or does not draw the line in
the usual place. This is the explanation I will be exploring. How-
ever,beforedoing thatlet me indicatethe pervasivenessof difficult
tense-sequences.I quote in leapfrogfashionto highlightthe verbs.
To regainmycomposureaftersuchlapsesI makea practiceof review-
ingmyburrow,and ... frequently leaveit.... It is alwayswitha certain
solemnitythat I approach the exit again ... [for] it was there that I
thispartof myburrow?I
began myburrow... Should I reconstruct
on
keep postponing the and
decision, the willprobably
labyrinth remain
as it is. ... SometimesI dream thatI have reconstructedit,... and now
... [The] nightson whichsuchdreamscometo meare
itis impregnable
the sweetestI know ... So I must thread the tormentingcomplications
... wheneverI go out ... Butthen(dann)I findmyself
of thislabyrinth
beneath the mossycovering [of the entrance] . . . and now (nun) only a
littlepush withmy head is needed and I am in the upper world. For a
long time I do not dare to make that movement ... I then cautiously
raise the trap door and slip outside . . . (pp. 331-333).
 M L N 561

The time of utterance of the first paragraph here is clearly the

same as at the beginning of the story: a present time after the
completion of the burrow, a point from which the creature looks
back to a cycle of habitual past behavior and forward to a future in
which the burrow will probably not be rebuilt. But again, when he
enters into closer description of his iterative excursions from the
burrow, the now of narration shifts and becomes the moment
(though what the status of that moment is we have yet to decide) at
which he leaves the burrow. This becomes particularly clear in the
paragraph that follows.
... I know .. . thatI do not have to hunt here (hier)forever,.. so I can
pass my time here quite withoutcare ... or rather I could, and yet I
cannot (vielmehr, ichkinntees undkannes dochnicht).My burrowtakesup
too much of my thoughts.I fled from the entrance fast enough, but
soon I am back at it again (schnellbin ichvomEingangfortgelaufen, bald
aberkomme ichzuriick).I seek out a good hiding place and keep watchon
the entrance... At such timesit is as if I were not so much lookingat my
house as at myselfsleeping ... In all mytime I have never seen anyone
investigatingthe actual door of my house . . . There have been happy
periods in which I could almost assure myselfthat the enmityof the
world toward me had ceased ... The burrow has probably protected
(schiitzt)me in more ways than I thought(gedachthabe) or dared think
while ... inside it.2... Sometimes I have been seized with (bekam)the
childishdesire never to returnto the burrow again, but to ... pass my
lifewatchingthe entrance.. . [But] whatdoes thisprotectionwhichI am
looking at here from the outside (die ichhierbeobachte) amount to . . . ?
. . . No, I do not watch over my own sleep, as I imagined; ratherit is I
who sleep, while the destroyerwatches ... And I leave my post of
observationand find I have had enough of this outside life.... But I
have never (nicht)been able to discover ... an infallible method of
descent. In consequence ... I have not yetsummoned the resolutionto
make my actual descent (ichbin ... nochnichtin den wirklichen Eingang
hinabgestiegen3), and am throwninto despair at the necessityof doing it
soon.... I tear myselffreefromall mydoubts and ... rush to the door,
... but I cannot.... The danger is by no means a fancifulone, but very
real ... If [an enemy] were actually to arrive now ..., if [it] were
actuallyto happen, so that at last ... I mightin my blind rage leap on
him [and] ... destroyhim . . . but above all-that is the main thing-
were [sic] at last back in my burrow once more, I would have it in my

2 The Muir translationreads: "... while I was inside it."

3 I followthe Brod texthere. The Pasley textis in error-cf. Henel, p. 23.
562 J.M. COETZEE

itselfwithrapture;butfirstI would... want

hearttogreetthelabyrinth
to rest ...But nobody comes . . . (pp. 334-337).
The tense sequence is itselflabyrinthine.The Muirs tryto follow
its twistingsand turnings,but there are unavoidable moments
when theyhave to choose between progressiveand nonprogressive
English forms(dieichhierbeobachte becomes "which I am looking at
here" ratherthan "which look at here") and between perfectand
I
preterite(binfortgelaufen becomes "fled" rather than "have fled").
There is no way,in fact,of translatingthe passage withoutcommit-
ting oneself from moment to moment to an interpretationof its
time-structure, in particularof the situationin timeof the moment
at which the narrator speaks: are the events beheld from the
perspectiveof the now of the firstsentence of the story-"I have
[now] completed the constructionof my burrow"-which would
make of the presenttense here a so-called historicalpresent,or has
the momentof narrationshifteddecisively,forthe timebeing, to a
time out in the freshair where the burrowingcreaturewaitsinde-
cisively,unable to venturethe descent back into the earth? In fact
this passage puts the question most starkly. "Ich bin . . . noch nicht
in den wirklichenEingang hinabgestiegen,"says the creature. If
the momentof utteranceof this sentence is the moment of utter-
ance of the text,then the creatureis nowliterallytrapped out in the
open.
This lengthyquotation should be enough to show that the de-
tailed progressionof tense-sequencesindeed raises puzzling prob-
lems. Withoutquoting at quite such length,let me point to further
passages in which the problem is unavoidable.
The creatureis "now" outside his burrow."For the present... I
am outside it, seeking some possibility of returning ..., con-
frontedby thatentranceover there (dort)which now (etzt) literally
locks and bars itselfagainst me" (pp. 339, 340). The deicticsem-
phaticallymark the momentof narrationas a momentoutside the
burrow. "And then ... I approach the entrance [and] ... slowly
descend" (p. 341). The nowof narratorialtimeshiftswiththenowof
narrated time: timeelapses in both the progress of the textand in
the world outside the entranceto the burrow,and "now" entrance
is achieved. The earlier irresolutionand incapacityto descend are
overcome by sheer exhaustion. "Only in this state [of exhaustion]
... can I achieve mydescent" (p. 341). But the returnto the burrow
rejuvenateshim. "It is as though at the momentwhen I set foot in
the burrow I had (hdtte)wakened from a long . . . sleep." He sets
 M L N 563

about transportingthe spoils of his hunting to his Castle Keep.

When thistask is completed "a feelingof lassitude overcomes me"
and he sleeps (pp. 342-3).
Though there is no break in Kafka's manuscriptat this point,4
there is a gap in narrated time."I musthave slept for a long time"
(Ich habe wohlsehrlangegeschlafen), the narrationcontinues.5This
second part of the storyconcerns the mysteriouswhistlingnoise
that the creaturehears in his burrow.Again the now of the narra-
tion seems to be cotemporal withthe now of the action; but again
there are unsettlingpassages in which the now seems to reveal an
iterativeface.6On the other hand, the noise is unambiguouslyde-
scribed as something"that I have never heard before" (was ichnie
gehirthabe) (p. 347)-an iterativereturnof the noise seems to be
ruled out.
When the firstresearches into the origin of the noise fail, the
creature revises his plans and speaks of a future of intention:"I
intend now to alter my methods. I shall dig a ... trench in the
directionof the noise" (p. 348). But thisnew plan bringsno solace,
for "I do not believe in it" (p. 349). The reason for thismistrustof
"reasonable" futureprojectionswould, in an iterativetime,be that
their failure has already been experienced. In the, so to speak,
blinkered present of the text the cause of his own hopelessness
remains obscure to the narrator.
Even if we read the entiresecond part of the storyas linear and
non-iterative,there are iterativecycleswithinit.
"SometimesI fancythatthe noisehas stopped...; sometimessucha
faintwhistlingescapesone . .. one thinksthatthewhistling
has stopped
forever.I listenno longer,I jump up, ..." (p. 350).
If, on the other hand, we read this part as iterative,then the se-
quence I have quoted becomes part of the iterativepresent: neither
German nor Englishwould appear to have a mechanismat the level
of structureof the verb phrase forindicatingiterativecycleswithin
cycles.
"It may happen (kann ... geschehen)that I (man) make a new

4 See Henel, p. 7.
5
Though the Muirs translatethe next fewverbsas preterites,theyare presentin
formin the German.
6 For
example: "In such cases as the presentit is usuallythe technicalproblem [of
trackingdown the noise] that attractsme" (p. 344); "often already I have fallen
asleep at my work" (p. 348); (when he begins to shovel soil) "this time everything
seems difficult"(p. 350).
564 J.M. COETZEE

discovery"(p. 351): thatthe noise is growinglouder. The shiftfrom

ich to man is maintainedfor much of the rest of the paragraph, in
conformitywith the new hypotheticalmode of the narrative. It
seems impossible to square this mode with a non-iterativeunder-
standing of the narrative unless one grants to the narrator the
effectiveposition of a fictionalcreator, someone toyingwith se-
quences whichmayor maynot be insertedintothe narrative.While
thispossibilitycannot be dismissedabsolutely,there is nothingelse
in the textto support the notion thatthe operations of writingare
being so radically unmasked. On the other hand, if one under-
stands the narrativeas iterative,then the hypotheticalsequence fits
in as one which may or may not occur in a given iteration.
As the creature moves about his burrow investigatingthe noise,
new ideas, new plans, new conclusions occur to him, all in turn
abandoned as useless. Why does he not rememberthem frompre-
vious iterations,why does he entertain them again if they are
proved ineffective,why does he experience surges of hope and
despair? The answer, at one level, is that he is in some sense con-
demnedto theseiterations,and thatpartof being condemned (as the
example of Sisyphusmightteach) is thatthe tormentsof hope are
part of the sentence. What should interestus particularlyin an
investigationof tense and time, however, is that the inabilityto
learn frompast failureis a reflectionof the factthatthe iterations
are not ordered:none of them being earlier in time than any other,
no iterationencompasses a memoryof an earlier one.
"Nothing ... approaching the present situationhas happened
before; neverthelessthere was an incidentnot unlike it when the
burrow was only beginning" (p. 355); and the creature digresses
into a past-tenseaccount of an episode fromhis "apprenticeship".
The temporal perspectivehas reverted unambiguouslyto that of
the opening of the story:a nowin the timeof narrationwitha linear
past behind it and a linear futurebefore it.
The last pages of "The Burrow",afterthisepisode, are resigned,
valedictoryin tone. The creature retiresto his Castle Keep, to his
storeof food, and awaits"the beast",dreaming of the peace of "the
old days" (p. 358). Perhaps it is possible that the beast has never
heard him, in which case there is hope. "But all remains un-
changed" (p. 359).

II
The extraordinarytime structureof "The Burrow" has been
commented on by numerous scholars. I should like to discuss two
of the more perceptiveof these commentaries.
 M L N 565

In her 1968 essay, "Kafka's Eternal Present",and again in her

book Transparent Minds (1978), Dorrit Cohn discusses peculiarities
of timeand tense in Kafka.7Since her commentson "The Burrow"
in the essay are absorbed into the book, I willquote only fromthe
latter.Cohn writes:
. . . The animal-midway through the story-seems to "forget" the
iterativenatureof his account and begins to tellof... the appearance of
the hissingsound. Up to thispoint the animal has described his habitual
subterranean existence in durative-iterativepresent tense.... [After
this point] the statictime of the firstpart of the story... becomes an
evolvingtime,its durative tense a punctual tense ... The speaker who
surveyedhis sovereignrealm in durative presenttense [is] transformed
into a monologistwho simultaneouslyexperiences bewilderingevents
and articulatesthem in a punctual present tense.
... This [temporal]structurecorresponds exactlyto Kafka's paradoxi-
cal conceptionof human time,whichis based on a denial of the distinc-
tion between repetitiousand singular events. For him, as he once af-
firmedaphoristically,"the decisive moment of human development is
everlasting"."The Burrow",by exploitingthe ambiguitiesof a discourse
cast in the presenttense, reflectsthis paradox in its language as well as
its meaning. If the crucial eventsof life happen not once, but everlast-
ingly,then the distinctionbetween durative and singulativemodes of
discourse is effaced: the durative silence always already contains the
hissing sound, and the destructionit brings lies not in a single future
moment,but in a constantlyrepeated present (pp. 195-7).
The discussionof pp. 334-7 of "The Burrow" above should make
it clear that Cohn's division of the storyinto a firstpart in which
"tense" is durative-iterativeand a second part in which it is punc-
tual, is too neat: shiftsoccur too frequentlyfor her generalization
to hold. Consequently, while she is right to characterize Kafka's
time-conceptionas "paradoxical [and] . . . based on a denial of the
distinctionbetween repetitiousand singular events",she goes too
far when she claims that this distinctionor opposition creates a
structure in any meaningfulsense. There is no clear correspondence
between,on the one hand, durative-iterative tenses and lifebefore
"the decisive moment"(the startof the hissing),and, on the other,
the arrivalof "the decisive moment"and singulativetenses. Where
Cohn does point in a fruitfuldirection is in identifyingthe am-
biguities of present-tenseverb forms as the formal field whose
exploitation makes the higher-levelparadoxes of "The Burrow"

7 "Kafka's Eternal Present: Narrative Tense in 'Ein Landarzt' and Other First-
Person Stories",PMLA, 83 (1968), 144-150; Transparent Minds (Princeton:Princeton
U.P., 1978).
566 J.M. COETZEE

possible. But there is a certain flaccidityin the argument that

Kafka's "denial of the distinctionbetween repetitiousand singular
events"is simply"reflected"in the language of the story.For "The
Burrow" does not "efface" the distinctionbetween "durative and
singulative".The mostwe can say is thatat certainpointsin the text
where we would expect the one formwe encounterthe other,and
vice versa. If the distinctionwere indeed effaced,and durativeand
singulativeformswere used interchangeably,the resultwould very
probably be nonsense. The problem is precisely that intuition
(whichmay mislead) suggeststhatthere is systembehind the aber-
rant usage; and our critical task is one of probing intuitionby
analysis.The conclusion I come to in thisessay happens to be quite
close to Cohn's: the storyis indeed dominated by "a constantly
repeated present."To reach thatconclusion,however,requires not
onlya tighterscrutinyof the textbut a principledunderstandingof
the use one may make of privilegedinsightssuch as the aphorism
of Kafka's that Cohn quotes.
In a study based on a more minute examination of tense se-
quences in "The Burrow" than Cohn's, Heinrich Henel arrivesat a
similarcharacterizationof the temporal situationof Kafka's crea-
ture: that it is "an endless condition".8Henel recognizes fromthe
startthe particularhermeneuticproblems posed by a textin which
so elementarya linguisticcategoryas tense, not easily reduced to
other terms,becomes the object of the writer'splay:
Whatkindof presentoccursat a givenpointis determined bytoneand
context;but whattone is appropriateand whatcontextis perceived
dependon howone understands thepresent(p. 5).
In Henel's reading, the storyfallsinto two main parts witha short
linkingmiddle passage. In the firstpart the use of the present is
indeterminate:
Oftenitsoundsas ifa uniquemomentin thehereand nowis intended,
yetthe dominantimpressionis of the iterative. . . Past definiteand
non-recurringeventsare reportedinthepreterite,butforthemostpart
earlierand nowmeltintoan endlesslyexpandingcondition(pp. 5-6).
In the second part
themeaningof thepresenttensechanges.Pastis clearlydistinguished
of thebeastproceedin
frompresent,and the thoughtsand activities
temporalorder... The narratornowkeepsstepwiththeeventsrepre-
8 Henel, p. 6.
 ML N 567

sented,and thepresenttensehe employsdenotesat each pointof the

narrationa different,
laterpresent.Whilethepresentof thefirstpart
fuseswithan untranscended past,thepresentof thesecondpartmoves
forward
consistently and mergesintoan indefinite future.The effectis
in bothcases thesame: an eternalconditionis represented(p. 6).
Thus, like Cohn, Henel is concerned to smooth out, by an act of
generalization,the difficulties presentedby the tense sequences. In
his reading the time of the firsthalf of the storyis, by and large,
iterative,the time of the second half is not. But, we can ask, is the
mode of generalizingfromthe totalityof data the correctmode of
argument to employ here? Are we concerned to formulatelaws
thatcover mostof the data-i.e., statisticalgeneralizations-or laws
thatexplain detailed variations,laws whose models would be rules
of grammar?9My aim in this essay is to elucidate the temporal
systemof the storyon the basis of usage which,despite its appear-
ance of aberrance, I must startby assuming to have some kind of
intentionalunity.For thisreason I do not find it enough to say, as
Henel does, that the present tense in "The Burrow" "fillsno less
than five distinctfunctions"(p. 4) without carryingthe analysis
further.10 This classificatory step is only a stage in analysis,withno
explanatorypower in The
itself. more importantstage is the one at
whichthe question is answered: Is there a coherenttime-system in
which these five functionscan be said to participate? In other
words: Is therea temporalcoherence to the story,or does the mind
behind the storyshiftfromone temporal subsystemto another?
9 The cases I cite in footnote6 above are enough to indicate that Henel's conclu-
sions are generalizationsratherthan laws,in the sense in which I use the terms.His
generalizationsare furtherweakened bya habitof selectivequotation. For example,
he writesof a "whollynew, hithertonever before grasped resolution"at which the
creaturearrives,"namlichvon dem Leben im Freien 'Abschied zu nehmen','niemals
mehr zuriickzukommen',und der 'sinnlosen Freiheit' auf immer den Riicken zu
kehren" (p. 6). The paradox Henel does not face here is that even this decisive-
sounding resolutionis given in a formwhollycompatible withan iterativetime,as
fullerquotation reveals: "Und ich habe Lust, Abschied zu nehmen . . . und niemals
mehr zuriickzukommen. . . GewiB, ein solcher EntschluBware eine vollige Narr-
heit, hervorgerufennur durch allzu langes Leben in der sinnlosen Freiheit" (pp.
121-2 in Pasley; pp. 335-6 in the Muir translation).It is the contentof the phrases
Henel quotes thatleads him to thinkof the resolutionas makinga break in the cycle;
but the paradox is preciselythatin thisstoryeveryirruptionintothe cyclesof timeis
so ambiguouslypresented in temporalformthat it seems at least capable of being
absorbed into the cycles.
10"As present proper, it describes an occurrence achieving itselfin the now; as
historicpresent an earlier occurrence; as iterativepresent a present occurrence
which has happened in the same way or a similarway fairlyoften; as progressive
present likewise a present occurrence which extends into an indefinite,perhaps
endlessfuture;and finallythe presentcan serveas a formof innermonologue"(p. 4).
568 J.M. COETZEE

III
HithertoI have used the word "tense"ratherlooselyto designate
the element of verb inflectionthat marks time-relations.I must
now refine the notion of tense by distinguishingbetween the two
elementsof verb inflectionwithtemporal functions:tense and as-
pect.
The theoryof the verb on which I shall be basing my discussion
of "The Burrow" is the descriptionfirstoutlined by Gustave Guil-
laume in Tempset verbe(1929) and subsequentlydeveloped in his
published lectures of 1948-9. A Guillaumean description of the
English verb systemhas been given by W. H. Hirtle.11(I am not
aware of any comparable studyfor the German verb.)
In Guillaume's theory,it is not possible to describe the systemof
tense and aspect in terms of a single model of time, namely the
familiarunidirectionalarrowof infinitetimeof Newtonianphysics.
The verb systeminstead rests upon two simultaneous and com-
plementaryways of conceivingtime: (a) as universetime,a limitless
linear time along whose axis any event can be situated; and (b) as
eventtime,the span of time that an event takes to achieve itself.
Though in theoryeventtimecan be infinitesimal, i.e., the eventcan
be purely punctual with no intervalbetween beginning and end,
this state is rarelyreached in the human world.12
Verbal aspectis a systemof representingevent time. Once this
mental representationhas been achieved, in Guillaume's theory,
the systemof tenseserves to combine the representationsof event
time and universe time.
How does aspect representevent time?It conceives of the event
as takingplace in two phases: a coming-to-be phase extendingover
successive instants,followed by a resultphase during which no
furtherdevelopment or actualizationof the event can take place.
Depending upon at whatpointof the temporalcontinuumthe verb
interceptsevent time, differentaspectual resultsare achieved. In
English,the primaryaspectual opposition is between (a) intercept-
ing event time at some instant(which may be the final instant)of
the coming-to-bephase, and (b) interceptingit during its after-

1 Gustave Guillaume, Tempset verbe(Paris: Champion, 1929); Leconsde linguis-

tique,I, ed. Roch Valin (Quebec: Laval, 1975); W. H. Hirtle, Time,Aspectand the
Verb(Quebec: Laval, 1975).
12 For discussion of this
point, see Bernard Comrie, Aspect(Cambridge: Cam-
bridge U.P., 1976), pp. 42-3.
 M L N 569

math. The two aspects which resultare, respectively,(a) immanent

and (b) transcendent.
A diagram (figure 1) may elucidate these concepts.

Interception: Interception:
immanent aspect transcendent aspect

B E
past non-past
Coming-to-be Result
phase phase

Figure 1

Here the continuum extending infinitelyfrom past to non-past

representsuniversetime;and the sectionBE representseventtime,
from beginning to end, with a coming-to-bephase and a result
phase. Depending on whethereventtimeis interceptedduring the
formeror the latterphase, we have verb formsof immanentaspect
("he is running","he runs", "he ran") or verb formsof transcen-
dent aspect ("he has run"). (We see fromtheseexamples thataspect
is independent of the past-presenttense distinction.)
How are iterativeverb forms-forms whose iterativemeaning is
signalled by non-syntacticmeans-represented in such a scheme?
Here the importantthingto recognize is that,though an iterative
form may be thought of as shorthand for a succession of single
events each with a beginning and an end (e.g., "he runs [every
day]"), it does not interceptthe resultphase of any of these single
events,and at most may or may not interceptonly the resultphase
of theirtotality.Thus in figure2, where each Bi and Ei represent
the beginningand end of a typicaliteratedevent i, the forms"he
runs/ran[everyday]","he is/wasrunning[everyday]", representan
interceptionin the coming-to-bephase of eventi, while "he used to
run [everyday]" representsan interceptionin the result phase of
the totalityof the iteratedevents,i.e., afterEn.

. . . B E .i En
1. E11
B1 E22
B22,a,1 ,./.-R.1 B.n

past B E non-past

Figure 2
570 J.M. COETZEE

Therefore withoutloss of generalitywe can condense figure 2 to

figure3, in whichthe iteratedeventsare representedwithoutindi-
vidual resultphases:

B1 B2 B3 *. Bi B.l * B

m 1E. 1E E.-E En

past B E non-past

Figure 3

There is one furtherpoint to recognize about iteration.Though

in the diagrams thus far I have represented the event that is iter-
ated as a single event (e.g. "I run"), in "The Burrow" it is more
oftena sequence of eventsof some length (e.g. "I must thread the
tormentingcomplicationsof this labyrinth.. . whenever I go out,
and I am both exasperated and touched when, as sometimeshap-
pens, I lose myself... But then I find myselfbeneath the mossy
covering... " [p. 333]). It is thewholeof thissequence of sub-events
(going out, threading, losing myself, being exasperated and
touched, findingmyself,.. .) which is iteratedand which is repre-
sented in figure3 by the event (Bi,Ei).
Now withinthe totaliteratedevent (Bi,Ei), morphologicalmeans
of time-specificationare more impoverished than under normal
(non-iterated)circumstances.This is because what is under normal
circumstancesa tense markerwitha secondary aspectual function
(e.g., the null morpheme 0 of run,which normallymarks the verb
as present in tense and only marks iterativeaspect when syntacti-
callyreinforced,as in "I run everyday"), is now charged primarily
with marking iterativeaspect, so that time-relationshave to be
specifiedby syntacticmeans. In the passage quoted above, the rela-
tive order of sub-eventsis represented (a) iconically,by the se-
quence of the signifiers("I lose myself... then I findmyself'),and
(b) by the logic of syntacticrelations ("I am ... exasperated ...
when ... I lose myself'),ratherthan morphologically.
This excursus on iterativity may help us to distinguishbetween
in
the structureof time "The Burrow" and the systemof tense and
in
aspect throughwhich, part, that structure is realized, and thus to
unravel, at least at a formal level, some of the complexitiesof the
narrative.For exemplarycloser analysis I have chosen the sequence
of pages in the firsthalf of the storybetweenthe emergence of the
 M L N 571

creature into the freshair (p. 333) and his descent back into the
burrow (p. 341), a passage in which the time-structure is perhaps
more bewilderingthan anywhereelse in the story.
If we scrutinizethese pages closely,we find an alternationbe-
tween two varietiesof temporal experience, and, going with each
variety,a particularnarrativepointof view.The ground-bassof the
passage is: (a) The iterativeexperience of emergingfromthe bur-
row, enduring the pleasures and terrorsof life above, not being
able to re-enter the burrow, then finally re-entering it. The
iterativityof the experience is signalled by so-called present-tense
(in fact iterative-aspect)verb forms with associated adverbials
(sometimes,usually,etc.). In figure3 the time-segmentof thisexpe-
rience is (Bi,Ei) and the moment of narration from which it is
described is outsideany (Bi,Ei), i.e., beyond E. But thereare regular
transitionsinto: (b) The timeof the iterationexperienced fromthe
inside, witha past and an unknown futureof its own. In termsof
figure3, it is as if the structureof (Bi,E) were identicalto that of
(B,E), and thereforeas though the iterativenature of the experi-
ence became invisibleor were erased from knowledge. There are
two formaldevices above all thatachieve transitionsof thiskind: (i)
the occurrenceof overtpast and futureverb-forms, whichhave the
effectof normalizingthe null morpheme 0 of the unmarked form
as a presenttenseratherthan an iterative aspectmarker(as, forexam-
ple, in the context he ran ... he will run, he runs is read as a
present-tenseform); (ii) the emphatic use of deicticslike now,this,
here,which,since theylocate the narrativerelativeto the time and
place of itsnarration,serve to introducethenowof narrationinside
(Bi,E,).
The movementof these pages is thus a continual slide from an
outside view of the cycle safety-danger-safety to an inside view in
which danger is experienced from the inside and from which it
seems impossibleto reattainsafety,followedby an abrupt and tem-
poraryreturnto the saferoutside view.This back-and-forthoccurs
not onlyat the level of the narrator'sexperience: it is also explicitly
thematizedin the passage as a "problem". It is possible to minimize
thisthematizationand read it as simplya privatejoke of Kafka's, a
wryreflectionon the experience of writingoneself into a corner.
But it is also possible to read it as a bringingto explicitnessof a
fundamentalexperience of time with which the storycontinually
wrestles at a formal level. Unable to summon the resolution to
re-enterhis burrow,the creature says: "For the present ... I am
572 J.M. COETZEE

outside it seeking some possibilityof returning,and for that the

necessarytechnicaldevices (technischenEinrichtungen)would be very
desirable" (p. 339). Among the mostdesirable devices would be, of
course, a passage fromthe dangers of (B,E) to the safetyof (Bi,EF)
(the switchingpower of the 0 tense/aspectmarkerwould be such a
device). Two pages later: "And then, too exhausted to be any
longer capable of thought,... I ... slowlydescend ... Only in this
state . . . can I achieve my descent" (p. 341). As long as conscious-
ness has been in control,thecreaturehas been unable to achieve this
transitionfromabove to below and has remained stuckin a condi-
tionthatis not onlyunendurable but logicallyimpossible:the itera-
tive formshave already promised that ascent and descent form a
cycle,thereforethe creaturecannot remain stuckhalfway.Exhaus-
tion and incapacityfor thoughtare the sole means that overcome
the arguments (or rationalizations)of the conscious mind which
keep him from his burrow; theyalso constitutethe absurd "tech-
nical device" that solves the problem of gettingstuck during the
cycle.What can be read in the mode of realismas a piece of rather
ineptdeusexmachinapsychologizingcan also be read in the mode of
text-construction as a flatteningof the distance between narrator
and narrated,tillthe adventuresof the creatureseeking a way into
his burrow become identicalwiththe adventuresof the signifying
subject seeking to find a way to keep the narrative moving. As
Henry Sussman writes:
The voice of the animal is ... also the voice of construction[of the
employedin
constructs
burrow,of thetext],thevoiceof therhetorical
thisparticular
production.
... The readeris askedtobelieveintheconcurrenceofthetextwiththe
actionswhichtheanimalclaimsto be performing at themoment.If for
no otherreasonthanbecausetheseactionsare mediatedby a written
text subject to time in differentways than the unidirectionalthrustof
is absurd.The narrativeconfinesitself,
experience,thispresumption
nevertheless,on the basis of this fictivetemporal immediacy,to a now
whichis remarkablyresistantto revisionsto the past or projectionsinto
the future.The animal thus becomes the agent of a temporalparadox,
thatthe now, capable of feedingupon itselfendlessly,is wider-reaching
than both the past . . . and the future. . .13

13"The All-EmbracingMetaphor: Reflectionson Kafka's 'The Burrow',"Glyph,1

(1977), 104, 106.
 M L N 573

Sussman is right to characterize time in "The Burrow" as

paradoxical. But the abilityof thenow to feed upon itselfendlessly
is not paradoxical at all as long as we distinguishbetween a now of
narrativetime (which tracksthe process of feeding) and a now of
narratedtime(thatwhichis fed upon). The paradox lies elsewhere:
in the apparent identity-if we rely upon the signals given by
verb-forms-of the texture of time in the narrated now of (BiE)
and the momentof narration.It is thisparadox whichKafka brings
into prominenceat the momentwhen, too "exhausted" to play any
longer withthe riddle itself,he cuts throughthe knotand puts the
creature back in the burrow.14

IV
It would be foolhardyto dismissout of hand the possibilitythat
"The Burrow" as we have it is incomplete, and that one of the
thingsKafka mighthave done if he had completed it to his own
satisfactionmighthave been to regularizeat least some of the more
bizarre tense sequences, or to create gaps in the text ("chapter-
breaks") to indicate lacunae in the time of narration.15Neverthe-
less, one's procedure as a criticmust be to test the possibilitythat

14 In the same part of his essay from which I quote, Sussman, however, gives a
characterizationof narrated time in the storythatignores the complexitiesof time
and aspect I have triedto outline,in particularthe "dissolve" from(B,E) to (Bi,Ei)
followed by reversion. Thus the followingargument of Sussman's, central to his
reading of the story,is so much the weaker:
In having recourse only to the here circumscribedby the constructionand the
now in which the work of constructiongoes on, or at least is contemplated,the
voice of the textabolishesthe "subject"whichis presumablyitssource and master.
Although the ruminationsof the animal are always in "self"-interest, in the ab-
sence of any subject,the self becomes the self of language, whose existence,like
the concept of the animal, defines the negation of the (human) self (pp. 104-5).
It is irrelevantforthe momentwhetherthe self is "the self of language" (Sussman's
thesis)or the self of narration(as I would prefer): all thatconcerns me here is that
Sussman's argumentis not well founded.
15 In the
postscriptto his editionof the story,Max Brod, on the authorityof Dora
Dymant,writesthatKafka completed"The Burrow",and thatin the pages lostfrom
the end the creature met his death in a fightwith his enemy. However, Heinz
Politzerargues cogentlythat there is no good reason to depend on Dora Dymant's
word and that the evidence points more stronglyto the conclusion that Kafka
himself destroyed the final pages, finding them unsatisfactory.See Max Brod,
"Nachwort",in Kafka, Gesammelte Schriften,V (New York: Schocken, 1946), p. 314;
Heinz Politzer,Franz Kafka: Parable and Paradox (Ithaca: Cornell, 1962), p. 330;
Henel, pp. 15-16. Kafka did not prepare the manuscriptfor publication.We may
thereforesuppose thatit lacks a finalrevision.
574 J.M. COETZEE

the textas itstandsis open to interpretation;only if no interpreta-

tion can be given should one fall back on the explanation that the
textis in some sense at fault.What I shall be doing in thissectionof
the essay, therefore,is to suggest how the repeatedly broken, in-
terruptediterativepresentcan be understood in the contextof the
whole of the story.
The state in which Kafka's creature lives is one of acute anxiety
(one would call it irrationalanxietyif there were any reliable op-
positionbetweenrationaland irrationalin his universe). His whole
life is organized around the burrow,his defense against an attack
which may come at any moment and withoutwarning. The key
notion here is withoutwarning.A warningis the sign of a transition
from peace to its opposite. Strictlyspeaking, the art of reading
warningsis purely prospective,future-directed:a sign recognized
retrospectively as havingbeen a warningis no longer a warning,for
it can no longer warn.
A warningis the sign of a transition.In "The Burrow",however,
time does not move through transitionphases. There is one mo-
mentand then thereis another moment;between them is simplya
break. No amount of watchfulnesswill reveal how one moment
becomes another; all we know is that the next moment happens.
Similarly,Zeno pointed out, before an arrow reaches its targetit
mustreach half-wayto itstarget;beforeit reaches half-way,it must
reach a quarter of the way; and so forth.To reach itstargetit must
pass throughan infinity of states;and to pass throughan infinity of
statesmust take an infinityof time. Zeno mighthave added: con-
ceiving the flightof an arrow in this way as a succession of mo-
ments,we can never understand how it gets fromone moment to
the next,we can never integrateits momentsinto a single flight.
We knowthatthisparadox (whichhe did not necessarilyarriveat
via Zeno) preoccupied Kafka. In "The Great Wall of China" he
describesthe messengerwho takes thousands of yearsand more to
bring a message from the Emperor. In "The Next Village" a
lifetimemay not be long enough for a journey to the next village.
In "Advocates" flightsof stairsexpand beneath the searcher'sfeet.
The mysticalcorrelate to the paradox is a time incommensurable
with human time in which man's life occupies a mere instant,yet
aeons of which can fitin the intersticesbetween two human mo-
ments.16
16
Cf. the notebook entryfor 11 December 1917, in which Kafka writesof the
momentof expulsion fromParadise as a momenteternallyrepeated, yetas belong-
ing to a time which "cannot exist in temporal relation" to human time.Gesammelte
Werke: aufdemLande(New York: Fischer/Schocken,
Hochzeitsvorbereitungen 1953),p. 94.
 M L N 575

Time in "The Burrow" is discontinuousin a strictlyformalizable

sense. Any momentmay mark the break betweenbefore and after.
Time is thus at everymomenta timeof crisis(fromGreek krino"to
separate, to divide"). Life consists in an attempt to anticipate a
danger which cannot be anticipated because it comes without
transition,withoutwarning. The experience of a time of crisis is
colored by anxiety.The task of building the burrow itselfrepre-
sents a life devoted to tryingto stillanxiety,naturallywithoutsuc-
cess; for withoutwarning "the enemy" is in the burrow. (Here I
suggestthatit would be naive to thinkthatthe whistlingis a warn-
ing and that"the enemy" is some beast whom the reader does not
get to see, ratheralong the lines thatDora Dymantsuggests;forby
the end of the storythe architectof the burrow clearlyrecognizes
thata break betweenbeforeand afterhas arrived,the clearestsign
of thisbeing that the lead-up time that once looked innocentnow
looks in retrospectlike a timeof warning[p. 355]. This does not of
course mean that there will only be a single foe, a single danger, a
single beforeand after:in theory"The Burrow" is infinitely exten-
sible.)
We treat the past as real insofar as present existence has been
conditionedor generated by it. The more indirectthe causal deri-
vation of the present froma particularpast becomes, the weaker
thatpast becomes, the more it sinkstowardsa dead past. But with
Kafka it is preciselythe power of each moment to condition the
next thatseems to be in question. Someone must have been telling
lies about JosefK., but no backwardexplorationof timewillreveal
the cause of the accusation against him. Gregor Samsa finds him-
self one morningtransformedinto a giant insect,whyand how he
will never know. Between the before and the after there is not
stage-by-stagedevelopment but a sudden transformation,Ver-
wandlung,metamorphosis.17
A common strategyof the first-person intelligenceattemptingto
understand the processesof timeis to takeup itsstancein a present
moment (ideally the moment of tranquillitywhen "I take up my
pen to write")whichstands forthe culminationof a certainpast, in
order to retracethe historythatled up to thismoment.Both parts
of Beckett'sMolloy,for example, take up this stance in an explicit

17I take the descriptionof the derivationalrelationshipof past to present from

Roman Ingarden, Timeand Modes ofBeing, trans. Helen R. Michejda (Springfield:
Thomas, 1964), p. 117. On the experience of the presentin Kafka, see furtherMax
Bense, Die TheorieKafkas (Koln/Berlin:Kiepenheuer & Witsch, 1952), p. 62; Jorg
Beat Honegger,Das PhdnomenderAngstbeiFranzKafka (Berlin: Schmidt,1975), pp.
29-31.
576 J.M. COETZEE

way.The firstsentenceof "The Burrow"seems to promisea similar

project: "I have completed the constructionof my burrow and it
seems to be successful"(p. 325). But the project soon turnsout to
be riddled with problems. Where are we to locate this privileged
momentof success and security:before,after,or during the recital
of eventsin and around the burrowthatoccupies pp. 325-343 and
terminatesin the state of sleep fromwhich the creature is awoken
by the whistling?As I have tried to show in section I of thisessay,
any putative temporal ordering of events at a detailed level be-
comes honeycombed with inconsistenciesand internal contradic-
tions.There is no smoothcourse of narrativedevelopmentthatwill
lead frombeginningsto the presentmomentof narration.Between
then and now is always a break.
It is fromthisvantage-pointthat the logic of iterativenarrative
becomes clear. Failing to trace the present to roots in the past,
Kafka'snarratorembarkson a seriesof projectsto wrap up the past
as a round of habitwhichincludes the presentand, insomuchas it
is repeated, projectsinto the future."I assemble mystores... I can
divide up my stores,walk about among them,play with them ...
That done, I can make mycalculationsand huntingplans ..." (p.
328): thisis typicalof the creature'sdiscourse.The crucial move, in
Guillaume's terms,is away fromuniverse time toward event time,
away from linear past-present-future tense organization toward a
cyclicaspectual organization of time.
This move-which I would call a ruse-is intended to capture the
relation of past to present to future by trapping them all in an
iterativepseudo-present.But as we have seen, the ruse continually
fails. The pseudo-present of iterative/habitual aspect continually
breaks down as the eventssignifiedwithin(Bi,E), the typicaltime-
gap, persist in organizing themselvesinto successivity,into time,
into tense, and then in collapsing in the persistentrupture of the
time-orderthat characterizes Kafka. There is no way of getting
here fromthere.
By talkingin termsof failed narrativeruses I may give the im-
pression thatKafka is in some sense against,above, and superior to
the narrator of "The Burrow", that if he does not know what a
successful narrativestrategymight be he is at least aware of the
futilityof the narrator'sstrategy.This picturewould entirelyfalsify
the story.V Thatwe have in "The Burrow",rather,is a struggle-not
onlythe representationof the strugglebut the struggleitself-with
timeexperienced as continual crisis,and experienced at a pitchof
 M L N 577

anxietythat leads to attemptsto tame it withwhatevermeans lan-

guage offers.The entire linguisticconstructcalled "The Burrow"
representsthe stillingof this anxiety;the major metaphor for the
linguisticconstructis the burrow itself,built by the labors of the
forehead (p. 328). But thisparticularburrow,"The Burrow",could
not have been built in a language that did not provide so easy a
means of glidingfromtense to aspect as German (or English) does.
Thus, withoutdenyingthe totalimplicationof Kafka in the story,it
should be possible to recognize that the particularformthe story
takes rests heavilyon a peculiarityof language. We can steer this
course withoutcommittingourselvesto the extremismof eitherthe
Whorfianthesisthatlinguisticstructuresdeterminethoughtor the
line characteristicof some Russian formaliststhatthe literarytextis
in some sense predeterminedby its devices.18
I can spell out my position in a differentway by isolating my
point of disagreementfromDorrit Cohn, whose Transparent Minds
contains the most carefullyworked out observationson the rela-
tions of time to narrativepoint of view in the story.Cohn recog-
nizes the "illogical"nature of its temporalstructure;but thisstruc-
ture, she says,
corresponds exactlyto Kafka'sparadoxicalconceptionof humantime,
whichis based on thedenialof thedistinction betweenrepetitiousand
singularevents.For him,as he once affirmed "thedeci-
aphoristically,
sivemomentof humandevelopment "The Burrow",by
is everlasting".
exploiting the ambiguitiesof a discourse castin the presenttense,re-
flectsthisparadoxin itslanguageas wellas itsmeaning.If thecrucial
eventsof lifehappen not once,but everlastingly, thenthe distinction
betweendurativeand singulativemodes of discourseis effaced:the
durativesilencealwaysalreadycontainsthehissingsound,and thede-
structionitbringsliesnotin a singlefuturemoment, butin a constantly
repeatedpresent(p. 197).
The aphorism Cohn quotes is both obscure and pregnant; but I
am not sure that it lends itselfto quite the point Cohn is making
here. It comes from the notebook of October, 1917, and occurs
aftera parable whose gist we mightexpress as follows: We die at
everymoment,but blindlydo not recognize our death and are spat

18
See, for example, Roman Jakobson: "It is the predominance of metonymy
which underlies and actually predeterminesthe so-called 'realistic'trend." Roman
Jakobsonand Morris Halle, Fundamentals ofLanguage (The Hague: Mouton, 1956),
(The Hague: Mouton, 1969), p. 195.
p. 78. See also VictorErlich,RussianFormalism
578 J.M. COETZEE

back into life. Kafka goes on: "From a certainpoint on, thereis no
more turningback. This is the point to be reached." And then:
"The decisive (entscheidende) moment of human development is
everlasting(immerwdhrend). Therefore those revolutionaryspiritual
movements that declare everythingbefore themselves null are
right,in that nothing has yet happened." The next aphorism is:
"Human historyis the second between two steps of a traveller."19
The passage as a whole thereforecontraststwo kinds of aware-
ness of time. The first,which we can call historicalawareness, im-
putes realityto a past whichit sees as continuous withthe present.
The second, which we can call eschatological,recognizes no such
continuity:thereis only the present,whichis alwayspresent,sepa-
rated from Ingarden's "dead past" by a moment of rupture, the
entscheidendeAugenblick.Hence the paradox thathistoryis over in "a
second" while the present moment is "everlasting".
To say, as Cohn does, that"the crucial eventsof lifehappen not
once, but everlastingly",thereforemisses the point. There are no
"crucial events" as opposed to other events: there is only what is
happening now, and thisis alwayscrucial.20Similarly,although the
linguisticopposition of durative to singulativecannot reallybe ef-
faced withoutcausing a general collapse of language, the concep-
tual opposition between the two-an opposition which belongs to
what I have loosely called the historicalsense of time-is brought

19Hochzeitsvorbereitungen, pp. 73-4. Quite aside from the literary-biographical

problem of relating the journal entryto a storywrittensome six years later, we
should be waryof erectinglarge interpretiveedificesupon journal entriesthatmay
be no more than fleeting,partial insightsdeveloped in greater precision by the
fiction.Cohn perhaps places too much reliance on this particular entry in her
reading of "The Burrow". On the qualities of the thoughtin Kafka'sjournals, see
Maurice Blanchot,"La Lecture de Kafka",inLa Partdufeu (Paris: Gallimard, 1949),
pp. 9-19. Blanchot writes:"TheJournal is full of remarksthat seem connected to
theoreticalknowledge . . . But these thoughts... relapse into an equivocal mode
that does not allow them to be understood either as the expression of a unique
happening or as the explicationof a universaltruth"(p. 10). Blanchot's essay con-
stitutesa caveat against abstractingKafka's thoughtfrom the particulardensityof
the experience it reflectson.
20 Cohn's
paraphrase would fitmore comfortablyover Kafka's meditations,in the
same notebook,on the eternal returnof the expulsion fromParadise (Hochzeitsvor-
bereitungen,p. 94); that is to say, they describe a mythicpresent. I would suggest
parentheticallythatpart of the reason forCohn's failureto push her conclusionsfar
enough maylie in her relianceon the treatmentof the presentin Harald Weinrich's
Tempus.Weinrichtreatsthe "historicpresent"as an "als ob" fora past timeand as a
component of a "Metaphorik der Tempora". It is, however, precisely the
metaphoricity of the narrativepresentthatKafka is bringingintodoubt in thisstory.
See Weinrich,Tempus:Besprochene und erzihlteWelt(Stuttgart:Kohlhammer, 1964),
pp. 125-9; Cohn, "Kafka's Eternal Present",p. 149.
 M L N 579

into doubt by a linguisticpractice that steps perilouslyalong the

brinkof contradiction,confusion,and nonsense. Thus by the end
of the storythe silence does indeed, as Cohn says, "alwayscontain
the hissing sound," and whateverthe noise signifiesis indeed al-
ready here "in a constantlyrepeated present" (which I would
rathercall an everlastingpresent).
But this does not go far enough. What is missingfrom Cohn's
account is a recognitionof the radical treatmentthatKafka gives to
narrativetime. For the everlastingpresent is nothingbut the mo-
mentof narrationitself.Now thatthe narratorhas failed timeand
again to domesticate time using strategies of narrative (i.e.,
strategiesbelonging to historicaltime), his structuresof sequence,
of cause and effect,collapsing each time at the "decisive moment"
of rupturewhen the past failsto run smoothlyintothe present,that
is, now that the constructof narrativetime has collapsed, there is
only the time of narrationleft,the shiftingnow withinwhich his
narrativetakes place, leaving behind it a wake (a text) of failure,
fantasy,sterile speculation: the ramificationsof a burrow whose
fatalprecariousnessis signalledby the whistlingthatcomes fromits
point(s) of rupture.
ofCape Town,SouthAfrica
University